Nies221

To get the size of archives and folders:
os.stat(filename).st_size # for folders you have to use it together with os.walk()

To get the list of files inside an archive I use the zipfile, rarfile and tarfile modules.
For the file type, I either extract the extension from the name or I use a bash command via subprocess:file --mime-type filepath
subprocess is used if there is no extension (as often happens with text files).

I preferred to use subprocess to avoid having to increase the number of dependencies needed.

.txt or plain text files without extension are supported.

instructions reference INSTALL.sh but that’s not in the source? Maybe only in the release tarball?

I added it to the repository: https://github.com/Nyre221/dolphin-quick-view/tree/main/package%20creation

does it clear the user’s entire clipboard history each time it’s run?

I made some changes and now it doesn't delete all the history, only the last thing copied (afterwards try to restore it but it doesn't work if you copied a folder or file): https://github.com/Nyre221/dolphin-quick-view/blob/main/package%20creation/quick_view_package/dolphin_quick_view_shortcut.sh

this is caused by dolphin's limitations and there's not much I can do about it.

would it make sense to package this up for PyPI, with system deps being checked for and reported from within python? If so, are you interested in pull requests?

I don't think it makes much sense for this application.
For now I'm trying to integrate the modules inside a .pyz and eliminate the use of pip: https://github.com/Nyre221/dolphin-quick-view/issues/10

Thanks!

I wonder how an implementation with KPart/C++ would look like

I'm not sure what you mean, I have no experience with c++.

Look, I've never used Apple's quick look or Gnome sushi so I can't answer this question. The people here can definitely give you a better answer than me haha

After reading about a user who was looking for an alternative to Apple's Quick Look and Sushi Gnome I decided to create this program to practice with python and make someone happy. I didn't expect there were so many people with the same need.

It looks absolutely great and I will download it, but I have a question, what advantage does it have, for example, to use this instead, for example, to open Okular to view a PDF?

I think its usefulness lies in the ability to scroll between files (even of different types) with the arrow keys, plus being fast.

Yes, I'm from r/cptsd.

Also searching for communities is hard without knowing about https://www.search-lemmy.com/

Strange, I found it very easily in the search results. Maybe your instance is hiding c/cptsd for some reason.

Right now the lemmy.world server is overloaded due to the influx of people from reddit. The situation should improve soon.

Still better than the official reddit app.

Welcome!

I’m here after leaving Reddit. R/raisedbyborderlines is the only place on Reddit I still visit.

The same for me.

I very much hope that many more people come to the fediverse. Reddit is pretty much an unpleasant experience on my phone and I much prefer a platform that isn't centralized.

That’s what makes it better than reddit. It can’t so easily be controlled by just a few people, because if one community/magazine on one instance gets overrun with toxicity, you can start a mag/comm with the same name on another instance.

absolutely, I am not against this kind of decentralization.

What I meant is that something could be done to collect the communities posts under one collection to make it easier for the user to join/see the communities content.

For example, if you subscribe to c/Technology you are subscribed to all the c/Technology communities (optional) in the federation. But this brings with it the problem of duplicates and I don't think there is an easy way to avoid it, and obviously different communities may have different rules.

Put simply: being able to subscribe to collections.

I say this because it seemed to me, at least initially, that for new users this presence of multiple communities with the same name was annoying and confusing.

Some time ago there was Zeronet (abandoned) that worked like this. The problem was that the speed of the site depends on how many users share the data and for heavy things like videos it was a catastrophe. On zeronet there was even a social network similar to facebook/twitter called ZeroMe: https://zeronet.io/docs/it/img/zerome.png

It was a really cool project, it's a shame it's not being developed anymore.